Vision-language-action

SmolVLA

SmolVLA is a compact base VLA exposed through LeRobot, with continuous action generation and a documented path to task-specific fine-tuning.

Source contextPrimary specReviewed 2026-09-28

Sources 2

SmolVLA base model card

Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

SmolVLA release

Hugging Face · Publication date not disclosed · Source accessed 2026-09-28

Reviewed recordLast reviewed 2026-09-282 sources ↗7 sourced fields

Official source Suggest a correction

Architecture and scope

The model card specifies multiview images, robot state and optional language instructions as inputs, with flow matching used for continuous actions. It explicitly describes the base model as a starting point for adaptation.

Source contextPrimary specReviewed 2026-09-28

Sources 2

SmolVLA base model card

Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

SmolVLA release

Hugging Face · Publication date not disclosed · Source accessed 2026-09-28

Integration and evaluation

The inference example connects a checkpoint, dataset preprocessing and an action postprocessor. Real robot evaluation uses a separately configured robot and camera setup.

Source contextPrimary specReviewed 2026-09-28

Sources 2

SmolVLA base model card

Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

SmolVLA release

Hugging Face · Publication date not disclosed · Source accessed 2026-09-28

Evidence boundary

Consumer-hardware positioning is a publisher claim; latency depends on the selected device and policy configuration.

Source contextPrimary specReviewed 2026-09-28

Sources 2

SmolVLA base model card

Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

SmolVLA release

Hugging Face · Publication date not disclosed · Source accessed 2026-09-28

Model identity

Developer: Hugging Face LeRobot
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
Exact version / scope: lerobot/smolvla_base
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
Category: Vision-language-action
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
Access model: Public model weights and LeRobot implementation
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
License: Apache-2.0 model card
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

Inputs, outputs & embodiment

Inputs: Multiview images, robot state and optional language
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
Outputs: Continuous actions
Primary specReviewed 2026-09-28
Sources 1
SmolVLA base model card
Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28
Embodiment: Not publicly disclosed

Implementation & training

Compute: Not publicly disclosed
Training-data disclosures: Not publicly disclosed
Integrations: Not publicly disclosed

Reported benchmark results

Context before scores. Results from different benchmarks are not directly comparable. A simulation result does not establish real-world reliability, safety or commercial availability.

No result meets our complete revision and methodology requirements for this record yet. Inspect the original benchmark documentation before comparing published scores.

What this evidence does not establish

Consumer-hardware positioning is a publisher claim; latency depends on the selected device and policy configuration.
Pricing and deployment service terms were not verified.

Relationships & deployments

Related reading is an editorial crosslink. Sourced connections describe relationships reported in the cited material. A link to a versioned profile does not establish compatibility with that version unless the connection note explicitly identifies it.

uses

SmolVLA → Hugging Face LeRobot

Sources 1

SmolVLA base model card

Hugging Face LeRobot · Publication date not disclosed · Source accessed 2026-09-28

Connection reviewed 2026-09-28

Curated comparisons

OpenVLA vs SmolVLA

Record history & verified changes

A research review records when we checked a source. It does not mark a product launch or a new deployment.

Initial reviewed record. No subsequent field change has been recorded.

Inspect the evidence

Sources & evidence

md-smolvla

SmolVLA base model card

Hugging Face LeRobot · Official spec · Publication date not disclosed · Source accessed 2026-09-28

License: Apache-2.0. Factual summary and attribution only; no upstream prose, images, weights, or dataset redistributed.

Referenced by 2 records: SmolVLA · Hugging Face LeRobot

md-smolvla-blog

SmolVLA release

Hugging Face · Official release · Publication date not disclosed · Source accessed 2026-09-28

Factual summary and attribution only; no upstream prose, images, weights, or dataset redistributed.

Referenced by 1 record: SmolVLA